INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iggs

-0.87

ural

-0.78

senal

-0.70

ership

-0.67

erva

-0.66

apse

-0.66

uggets

-0.65

duction

-0.64

justice

-0.64

algia

-0.62

POSITIVE LOGITS

do

0.72

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.71

ãĤ¦ãĤ¹

0.71

 Adin

0.68

bart

0.68

sided

0.67

SourceFile

0.66

 poss

0.64

Versions

0.64

){

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.